Segmentation of the vertebrate anterior-posterior axis is dependent on oscillations of her/he's genes. In humans, mutation of HES7 results in spondylocostal dysostosis, a congenital disorder characterized by fusion or malformation of the vertebrae and ribs. Within the zebrafish segmentation clock, six members of the her/hes family form both hetero- and homodimers and negatively regulate their own transcription. Even within this simple network, there are 36 possible dimer-DNA interactions and, given 3 levels of repression, 1017 possible network topologies. Previous models of gene networks based on transcription factor (TF) binding included data on binding affinities, cell-specific protein concentrations and chromatin accessibility. However, TF binding may not always lead to significant changes in transcription and in many cases biochemical data may be incomplete. Here, a method for determining transcription factor networks that does not need extensive biochemical datasets will be developed. Mapping the network will rely on gene knockdown and quantitative measurement of transcription combined with mathematical modeling and global optimization algorithms. Predictions derived from the network model will be tested experimentally via compound gene knockdown, qPCR and in situ hybridization. Network-based explanations for several incongruous phenotypes will be attained. Lastly, modeling these interactions will be used to shed light on how networks can evolve between species or within an organism to complete a different task with minimal modifications. Understanding gene networks, such as the segmentation clock studied here, enables modeling of biological processes. These models can lead to identification of nodes of cellular pathways, the role of multiple genes in disease phenotypes, and insight into drug interactions.